In genetic disorders associated with premature neuronal death, symptoms may
not appear for years or decades. This delay in clinical onset is often ass
umed to reflect the occurrence of age-dependent cumulative damage(1-6). For
example, it has been suggested that oxidative stress disrupts metabolism i
n neurological degenerative disorders by the cumulative damage of essential
macromolecules(1,4,7). A prediction of the cumulative damage hypothesis is
that the probability of cell death will increase over time. Here we show i
n contrast that the kinetics of neuronal death in 12 models of photorecepto
r degeneration, hippocampal neurons undergoing excitotoxic cell death(8), a
mouse model of cerebellar degeneration(9) and Parkinson's(10) and Huntingt
on's diseases are all exponential and better explained by mathematical mode
ls in which the risk of cell death remains constant or decreases exponentia
lly with age. These kinetics argue against the cumulative damage hypothesis
; instead, the time of death of any neuron is random. Our findings are most
simply accommodated by a 'one-hit' biochemical model in which mutation imp
oses a mutant steady state on the neuron and a single event randomly initia
tes cell death. This model appears to be common to many forms of neurodegen
eration and has implications for therapeutic strategies.